URL tools send a public HTTP or HTTPS address to SolViewer’s fetch service. The service accepts only those two schemes, rejects URLs that carry usernames or passwords, and refuses localhost, .local, .internal, and private-network addresses before it opens a connection.
Each request waits at most 12 seconds, follows at most five redirects, and reads at most 2 MiB of the response (a sitemap request may read up to 50 MiB; a child sitemap or robots.txt request stays at 2 MiB). The fetch does not run page JavaScript, so markup injected only after the document loads is absent from the result. Discovery probes used by the SEO Checker look for /robots.txt, /sitemap.xml, /llms.txt, /llms-full.txt, and /favicon.ico on the same origin; those probes are samples, not a site crawl.
Pasted HTML, pasted robots.txt, pasted headings, and the HTML Viewer stay in your browser. They are not sent to the fetch service.